CD10/Neprilysin as an Integrated Transcriptional Readout of Alveolar Macrophage Homeostatic Identity: A Multi-Dataset Secondary Analysis

Background. Alveolar macrophages (AMs) are long-lived resident sentinels of the alveolar lumen. Yet no single surface readout coupled to their homeostatic program, with an explicit interpretation rule in disease, has been consolidated. CD10 (MME/neprilysin) is listed among human AM surface markers, but its performance as an identity readout and its biology have not been characterized across datasets, programs and disease states. We asked whether CD10 marks AM homeostatic identity, which programs it couples to, and how it changes in disease. Methods. We performed a secondary analysis of the Madissoon healthy human lung atlas and the Natri lung-disease atlas, with bronchoalveolar lavage (BAL) studies as literature-level reference. We quantified MME detection across compartments, co-expression programs, MME+/MME− dichotomized comparisons with sequencing-depth-matched controls, and donor-level disease statistics using the donor as the statistical unit. Results. CD10 marked resident AMs: MME was detected in 30.1% of AMs and 51.9% of proliferating AMs, versus 0.2–0.3% of monocytes. Within AMs, higher MME tracked the PPARγ lipid and TGFβ maturation programs (Pearson ρ = 0.296 with ITGB8, ρ = 0.209 with PPARG) together with a don’t-eat-me profile (3.20-fold PD-L1, 1.39-fold CD47); these couplings survived sequencing-depth-matched controls. At donor level, the MME–FABP4 coupling was reproduced (ρ = 0.472) whereas MME–PPARG was not. AM-compartment MME did not differ between idiopathic pulmonary fibrosis (IPF) and control (p = 0.92; 36 versus 44 donors), and no AM-compartment change survived FDR control in any diagnosis, whereas interstitial macrophage MME increased in IPF (p = 0.0003, FDR = 0.003) and non-IPF interstitial lung disease (p = 0.0042, FDR = 0.023). Conclusions. CD10 is a transcript-level candidate readout of AM homeostatic identity, co-expressed with the PPARγ lipid program and elevated don’t-eat-me signals. Disease-associated changes are compartment-specific, stable within AMs and induced within interstitial macrophages, consistent with compartment remodelling rather than wholesale loss; measured on tissue composition, dilution alone predicts that AM-dependent samples lose roughly half of their CD10 signal in fibrosis, quantitatively matching prior reports of resident AM depletion. Protein-level validation is pending.